Backend v3

The SciCat backend v3 is the SciCat metadata catalogue RESTful API layer, built on top of the Loopback framework.

Configuration options

The v3 backend configuration is set through files. What follows is a list of available options. Some configurations are very verbose and could be simplified, but v3 has reached end of life in favour of v4, thus there is no active development.

datasources.json

It allows setting the connection to the underlying mongo database. It consists of two blocks: It consists of two blocks, the transient one which should not be changed and the mongo one for which we list the options that can be configured.

mongo

TL;DR in most cases, is enough to set the desired url with the full connection string to your mongo instance.

Name	Description	Value
host	mongodb host	"mongodb"
port	mongodb host port	"27017"
url	mongodb full URL. If set, all the other options, apart from useNewUrlParser and allowExtendedOperators are discarded in favour of this one	""
database	mongodb database	"dacat"
password	mongodb user password	""
user	mongodb user password	""

providers.json

It allows setting the authentication providers. The local block sets the local accounts.

Any file called providers*.json will be merged together by the merge_json.sh . This is done to allow better scoping of providers options.

local

The only option available is to either enable or disable the local authentication. Remove the block if you want to disable that.

config.local.js

It allows setting backend-specific configurations. Here are the commonly changed options.

exports

Name	Description	Value
pidPrefix	prefix of the internal IDs	"PID.SAMPLE.PREFIX"
doiPrefix	prefix of the published DOIs	"DOI.SAMPLE.PREFIX"
policyPublicationShiftInYears	number of years before the data should be made open access. This is only an annotation in the metadata and no action is triggered after expiration	3
policyRetentionShiftInYears	number of years by which the data should be kept. This is only an annotation in the metadata and no action is triggered after expiration	10
site	name of the facility runnin SciCat	"SAMPLE-SITE"
queue	message broker flavour for the JOBs	"rabbitmq"
logbook.enabled	option to enable scichat	"false"

Functional Accounts

There are a few functional accounts available for handling data:

Username	Password	Usage
admin	2jf70TPNZsS	Admin
ingestor	aman	Ingest datasets
archiveManager	aman	Manage archiving of datasets
proposalIngestor	aman	Ingest proposals

Default configuration

In the default configuration folder config , the backend is set to use the mongo container .

Enable additional features

Additionally, by setting the env variable JOBS_ENABLED , the archive mock and rabbitmq services are started and the backend is configured to connect to them.

If LDAP_ENABLED is toggled, you can use LDAP to log in with a LDAP user .

If OIDC_ENABLED is toggled, you can use OIDC to log in with a OIDC user .

With DEV=true , since the v3 tests are supposed to run with an empty DB, the set DB is dacat_test which is empty. If willing to use the seeded one later during development, just set dacat as database values in the file /home/node/app/server/datasources.json on the container.

Dependencies

Here below we show the internal dependencies of the service, which are not already covered here and here (if B depends on A , then we visualize as A --> B ). The same subdomain to service convention applies.

graph TD
    rabbitmq --> archivemock
    rabbitmq --> backend
    backend --> archivemock